A hospital-based case-control study of companion dogs examined the risk of developing canine malignant lymphoma associated with the use of chemicals in and about the home. Information from a self-administered owner questionnaire and/or a telephone interview of about 491 cases, 466 nontumor controls, and 479 tumor controls indicated that owners in households with dogs that developed malignant lymphoma applied 2,4-dichlorophenoxyacetic acid (2,4-D) herbicides to their lawn and/or employed commercial lawn care companies to treat their yard significantly more frequently than control owners (odds ratio = 1.3). In addition, the risk of canine malignant lymphoma rose to a twofold excess with four or more yearly owner applications of 2,4-D. The findings in this study are consistent with occupational studies in humans, which have reported modest associations between agricultural exposure to 2,4-D and increased risk of non-Hodgkin's lymphoma, the histology and epidemiology of which are similar to those of canine malignant lymphoma. The present study suggests that human health implications of 2,4-D exposure in the home environment should receive further investigation.

Histologically confirmed testicular tumors were diagnosed in 410 dogs from 12 North American veterinary university hospitals and clinics. Three tumor-cell types, Sertoli cell tumor, interstitial cell tumor and seminoma, were about equally represented. Several breeds were identified with high risk for different testicular tumor-cell types. Cytogenetic and immunogenetic studies of these dog families could offer leads applicable to familial testicular cancer in man. The multiplicity of breeds within the series suggests that, as in man, other factors, in addition to hereditary, play a role in etiology. Cryptorchid dogs appear to have a 13.6 times higher risk of testicular tumor than normal dogs. Additionally, male dogs with an inguinal hernia have an increased risk (4.7) of testis tumors. There were no detectable excesses of other urogenital anomalies or urogenital tumors among the series. The Shetland Sheepdog is suggested as an appropriate model for research into the mechanisms responsible for testicular maldescent and tumorigenesis.

A retrospective study of 2,912 cryptorchid dogs identified 14 breeds with significantly high risk. Among six distinct closely interrelated breed groups (e.g., toy, miniature, and standard poodles), the risk in the smaller breed was always greater than that in the larger relative, suggesting that genetically influenced maldescent could be, in part, related to physical size or the rate of growth of the involved structures. Testicular tumors were diagnosed in 5.7% of the cryptorchid dogs; half had only Sertoli cell tumors, one-third had only seminomas. The relative risk for Sertoli cell tumor or seminoma was not directly related to a familial risk for cryptorchism. Using the health experience of a control population composed of male dogs with anal sac disease (N = 4,184), there is an estimated relative risk of 9.2 in cryptorchid dogs to develop a testis tumor (95% confidence interval, 5.9-14.3) and 4.2 in dogs with inguinal hernia (95% confidence interval, 1.8-9.5). Considering that the anatomical development of the genital tract, testis descent, and tunic relationships in dog are very similar to that in man, and that the associations of cryptorchism and inguinal hernia with testis neoplasms are also similar, the dog should be an excellent model system to further investigate the causes of human cryptorchism.

From 11 North American veterinary university hospitals and clinics, 248 animals were a confirmed diagnosis of nervous-tissue tumor were identified; 7 tumors were found in cattle, 28 in horses, 14 in cats, 199 in dogs, and none in other species. Tumors were divided for analysis into three categories-glial, meningeal, and peripheral nerve. In cattle and horses, all tumors involved peripheral nerves, the risk of which, in horses, reached a plateau at 4-6 years of age and remained constant thereafter. In cats, the tumors were equally distributed among the three tumor categories whereas, in dogs, twice as many glial tumors as meningeal and peripheral nerve tumors were found. The risk for glial tumors in dogs reached a peak at 10-14 years of age, for meningeal at 7-9 years, and for peripheral nerve at 2-3 and 7-9 years. Three canine breeds-English bulldog, boxer, and Boston terrier-had an excessive rish of glial tumors. Except for an excess of skin tumors in dogs with peripheral nerve tumors, there was no unusual occurrence with second primary neoplasms for any species. There was no detectable predisposition by sex for any of the categories of nervous-tissue tumors among any of the four species. The role of genetic abnormalities associated with nervous-tissue tumors and other etiologic factors (e.g., chronic hypoxia) may be clarified by further studies involving canine breeds of "bulldog" ancestry.

Causes of death among 5016 white male veterinarians identified from obituary listings in the Journal of the American Veterinary Medical Association were compared to a distribution based on the general US population. Proportions of deaths were significantly elevated for cancers of the lymphatic and haematopoietic system, colon, brain, and skin. Fewer deaths were observed than expected for cancers of the stomach and lung. Although socio-economic and methodological factors may be involved, the patterns suggest that sunlight exposure is responsible for the excess of skin cancer among veterinarians whose practices are not exclusively limited to small animals, and ionizing radiation exposure contributes to the excess of leukaemia among veterinarians practising during years when diagnostic radiology became widely used. Mortality was also high for motor vehicle accidents and suicides, but low for diseases of the respiratory system.